1. Field of the Invention
The present invention relates to the field of wireless communications. More particularly, the present invention relates to the combined practices of Network Performance Management and Mobile Device Management. Even more particularly, the present invention relates to one or more mobile hosts monitoring one or more data measurements of one or more public wireless networks from one or more terminal nodes of a network, locally storing the collected data, processing the data through an artificial intelligence engine, and periodically communicating the collected data back to a centralized data collection server where it may again be processed through an artificial intelligence engine and stored into a database so it can be viewed with a graphical and analytical front-end user interface.
2. Discussion of Background Information
Within the last two decades, wireless networks and the surrounding ecosystem of mobile computing products have been steadily gaining in market adoption. The promises of wireless adoption include high return on investment, increased mobile worker productivity, ubiquitous public wide area networks, high network speed, and high network security. In many cases, these promises have been realized. However, in many other cases, the value gained from wireless adoption has fallen short of expectations.
For many enterprises, deployment of mobile solutions and adoption of public wireless networks have included problems such as unexpected overages and fees, dropped calls, lost connections, intermittent coverage, lower than anticipated bandwidth per mobile worker, and variations in network trust. In addition, current trends in public wireless network supply versus demand are expected to drive the elimination of unlimited pricing plans in favor of tiered pricing plans with specified usage limits. These trends will serve to exacerbate the pain currently felt by enterprises trying to manage and control expenses related to their mobile workforce. In addition, with the increased adoption of wireless networks and an increasingly mobile workforce, as well as trends in mobile broadband technology development, enterprises have found that mobile assets are fundamentally more difficult to manage than fixed assets.
Historically, enterprises have turned to network performance management tools to help control the problems listed above. Unfortunately, most existing products in the marketplace were designed for wired networks and for wireless networks that are fully controlled by the enterprise (ie private WiFi among others).
Most of the existing products in the marketplace gather performance data on the networks using data collection agents in the network infrastructure (ie routers, switches, among others). When the infrastructure is inaccessible to the enterprise, because the network is public, these tools do not work. In addition, many of the existing products in the marketplace communicate collected data back to a central server using standard protocols such as Simple Network Management Protocol (SNMP) or Netflow. While these protocols work well on traditional wired networks, they are chatty, inefficient, and result in inflated network usage costs when used on public wireless networks.
In addition standard systems developed for wired networks rely on snap shots of data being available to build historical knowledge of how a systems state varies over time. For example, Simple Network Management Protocol (SNMP) will continually poll a device for network statistics taking a temporal snap shot of the state of the Transmission Control Protocol (TCP) stack at the instant of each poll. This snapshot of data is then stored on a server for future analysis. These standards based management systems have gaps in knowledge created by intermittent connectivity when running over wireless networks due to regions of low signal and connectivity errors. If a mobile device is unable to connect when a sample is requested by an SNMP management system the mobile device's state at that instant and location is lost forever and cannot be used for future analysis.
Another example demonstrating the limitations in the current state of the art for network management systems is RFC 3954 Cisco Systems NetFlow Services Export Version 9, and in particular to section “3.3—Transport Protocol.” The disclosure of RFC 3954 is expressly incorporated by reference herein in its entirety. The system is designed without regard for congestion—let alone intermittent connectivity. RFC 3954 recommends a dedicated link from the data collection agent to the server specifically to avoid solving the congestion or intermittent connectivity problems. This type of system obviously cannot allow an enterprise to manage their use of public wireless networks.
Also, the performance characteristics of wireless networks are unique from wired networks in that they vary over space and time. Two points, separated by space, can and often will experience differing levels of network quality on a wireless network. Further, measurements of network quality on a wireless network for a single point in space but with measurements separated in time often vary as well. Traditional network performance management systems do not collect Geographical/Geospatial Information System (GIS) location as the data collection agents are deployed to network infrastructure hosts that are fixed in space. Traditional systems do not account for network measurements collected over time and correlated to a dynamic GIS location.
Therefore, a need exists for enterprises to collect data about devices using public wireless networks and the network performance that the device experienced over time correlated to device GIS location so that enterprises can determine problematic devices and so that enterprises can determine problematic areas of the public wireless network. Additionally a method is needed to maintain the historical knowledge of how a device's state (location, packet counts, signal strength, running applications, processes, errors etc) changes over time even when the device is in areas of poor signal strength preventing a good connection or simply is intermittently connected to a network so that historical trends can be monitored without loss of information. Additionally, a need exists to collect data about devices using public wireless networks and their network usage levels over time and location so that enterprises can control costs associated with excess usage or costs associated with under-used devices. Additionally, a need exists to collect data about highly mobile devices equipment inventories and usage patterns so that enterprises can better manage mobile assets. Additionally, a need exists to minimize bandwidth requirements of transmitting collected data to a central server so as to minimize usage cost overhead of doing so. Additionally, a need exists to facilitate analysis of the collected data to ease the burden of the above mentioned problems by making all collected data accessible in a graphical front end reporting system that provides GIS map and chart based visualizations of the correlations among the collected data.